Departmental Bulletin Paper 解探索の広域化を図る改良粒子群最適化アルゴリズムの提案及び性能評価

松村, 修平

The optimization problems for solving the objective function to the maximum or minimum under the given constraint condition are important problems applied to various fields. Metaheuristics is used as a methodto obtain an acceptable solution within the practical time for the problems and Particle Swarm Optimization (PSO) is one of them. The characteristic of PSO is it has superior convergence speed, easy implementation oncomputer and there are many application examples. However, as with other methods, depending on the problems, it may be difficult to obtain a global optimum solution by fitting to local solutions.In this paper, we propose improved PSO using exchange of particles’ position vector, re-initialization and movement constraints in order to widen its search range and show the effectiveness through experiments.Key Words : Particle Swarm Optimization, PSO, Optimization problems

Number of accesses :  

Other information